PMITS06

On this project, the main focus resides on the research of methods and algorithms for robust and reliable perception of pedestrians and other moving obstacles in ITS specific scenarios. The perception system will integrate information from both on-board and off-board tracking and classification modules. These modules are the research core of this proposal together with the fusion module that will integrate them. The project aims at enhancing the perception capability of an ITS, pursuing the following scientific and technical goals: Improve object classification algorithms and design new ones using several sensing modalities (laser, ultrasound, vision) gathered from on-board and off-board sensors; Algorithms for multi-view detection and classification of objects, pedestrians and vehicles using a calibrated camera network; Algorithms to robustly fuse information provided by object tracking and classification modules; Analyze algorithms for object detection and tracking modules and adapt them to come up with a context-based object tracker (well adapted to each specific situation); Information fusion techniques that will allow to obtain accurate detection and localization of vehicles and pedestrians by making use of data obtained both on-board and off-board the vehicles.